© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 3072
    9. 8-RES_FS3072-JB
    11. 759
    Prev
    Next
    INDEX
    Explanations

    phrases related to computer programming concepts and financial terms

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.14
    0.6%
    481
    +0.12
    0.5%
    457
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.19
    616
    +0.12
    0.09
    680
    +0.10
    0.07
    Negative Logits
    !.
    -0.84
    $.
    -0.81
    `.
    -0.80
    ilaterally
    -0.74
    *.
    -0.73
    utics
    -0.73
    ecause
    -0.72
    +.
    -0.72
    Ú
    -0.71
    outube
    -0.70
    POSITIVE LOGITS
     portion
    0.88
     analogy
    0.84
     consisted
    0.81
     shortage
    0.81
     consists
    0.80
     version
    0.80
     approach
    0.79
     question
    0.79
     gap
    0.77
     thing
    0.76
    Activations Density 2.996%

    No Known Activations